ThalisAI/Qwen2.5-72B-Instruct-heretic
Full fine-tune vs Qwen/Qwen2.5-72B-Instruct. ThalisAI/Qwen2.5-72B-Instruct-heretic is a GGUF (Q8_0) of different weights. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 0/161, MLP 240/240, Attention 320/560, lm_head 1/1